JFC (Jefsey) Morfin jefsey at
Fri Feb 18 15:57:48 CET 2005

At 13:06 18/02/2005, Michael Everson wrote:
>I'm not a programmer, and I don't think like one, and I don't live and 
>work in an environment where this kind of vocabulary use makes sense or 
>where anyone uses it this way. (I don't know what a heuristic is, either.)

Dear Michael,
Sorry for what I said to you. This explains a lot: you are just an outsider 
to the programmer's world of which you are the supposed to be the pilot for 
the most sensitive societal issue of the Information Society. You obviously 
need help rather than flames.

PS. For the information of those who will read this mail in public archives 
( :

Michael Everson is the person the IESG assigned the mission to decide about 
the entries and updates in the Internet Protocols Parameters clearing house 
(the IANA). These entries are to give programmers, developers, computers 
and applications the golden bytes they need to know what to do with the 
data they are to process at the three "language" layers 
(internationalization of the protocols, multilingualization of the 
exchanges, vernacularization of the interfaces).

Due to the current state of  the IAB/IESG/IETF thinking, Michael Everson is 
to decide for the entire Internet world, de facto for the whole digital 
world and by consequence for most of the cultural world, what a given 
language is and who is documenting it. He does that in filling an exclusive 
IANA slot. This slot is named in a computer  way: in using a tag (the 
"langtag"). This tag is a made of a unique language+script+nation 
combination. It is built in using codes from ISO tables which may differ 
from real life IANA other codes, resulting in confusion in application and 
Internet procedures. For example, for archives access reasons ".su" has 
been retained for the USSR related sites, however "ru-Cyrilic-SU" (Sovietic 
Russian) cannot be documented.

There is one single fr-Latn-FR slot and therefore one single language 
vision/version permitted for the French of France. There is no en-Latn-UK 
for the English being used in relation to the ".uk" namespace (the UK code 
is to be "GB"). This unique exclusive slot per language/script/country does 
not permit to accommodate any other lingual, academic, cultural, historic, 
community, political, technical, legal, etc. semantic authoritative 
considerations nor dictionaries than the single one Michael Everson is to 
chose. It does not permit either to accommodate any other vernacular style 
in the use of a languages that the one he is to select.

Complex sub-tagging is possible to address such "details", but it is likely 
that the resulting programming constraints will either greatly limit their 
development or lead to protected proprietary solutions. The current rhythm 
of registration of a language tag is at best one or two per day, if 
disputes are limited (this has however not started). There are 7260 
languages using 1 to 3 scripts in 250 political areas. One therefore can 
expect the need for 100.000 langtags. The need (qualified as urgent by the 
W3C) could therefore be addressed in March 2205 (without taking into 
account the more complex maintenance aspects), more likely 2505, 3005 if 
maintenance is to be considered due to the extensive research being needed.

This partly results from the disinterest of IETF in lingual issues and from 
the inability of Michael Everson (probably for the reason he gives) to 
understand and explain the IESG that:
- the langtag combination must also include an authoritative source and a 
style descriptors when documenting the way a computer is to run a language 
oriented process,
- the system must permit defaults for more general visions of a language 
and language tags synonyms when different semantics use different codes for 
the same description.
- signed tags permit to empower lingual authorities and for them to have a 
declarative/retrieval approach much like for Domain Names 
registration/whois, for a meta 
registry in immediate operations after a short sunrise period.

Current language registrations under considerations for languages such as 
Chinese, Tajik, Mongol, Bosnian, Inuktitut, etc. are introduced and 
documented by IBM and Microsoft without report of contacts with the 
concerned ccTLD Manager, Government Cultural authorities, Academic world, 
Publishers, Press, Media, Artists associations, Content Providers, Civil 
Society representative or UNESCO programs.

If you wish to subscribe to the open mailing list were registrations are 
reviewed prior to the approval of Michael Everson (along with the procedure 
defined by RFC 3066 - . Only a larger 
number of Internet competent persons on this list will help Michael Everson 
understand his IANA world is a world of programmers or of non-programmers 
trusting his programming culture understanding and propose a better way for 
the world to deal with the responsibility of the digital multilingualism 
than to delegate it to a brilliant linguist having no programming culture 
nor understanding.

More information about the Ietf-languages mailing list